Skip to content

Instantly share code, notes, and snippets.

View bushaev's full-sized avatar

Vitaly Bushaev bushaev

  • ITMO University
  • Saint Petersburg
View GitHub Profile
@bushaev
bushaev / rprop.py
Created September 1, 2018 17:50
Siplest form of rprop algorithms
for t in range(num_interations):
dw[t] = compute_gradient(x, y)
if dw[t] * dw[t - 1] > 0:
step_size = min(step_size * incFactor, step_size_max)
elif dw[t] * dw[t - 1] < 0:
step_size = max(step_size * decFactor, step_size_min)
w[t] = w[t - 1] - sign(dw[t]) * step_size
for _ in iterations:
hello_world = "Hello World"
for _ in iterations:
hello_world = "Hello World"
for _ in iterations:
hello_world = "Hello World"